Methylation-independent Binding to Histone H3 andCell Cycle-dependent Incorporation ofHP1� into Heterochromatin*□S �

Received for publication, January 19, 2006, and in revised form, March 13, 2006 Published, JBC Papers in Press, March 18, 2006, DOI 10.1074/jbc.M600558200

George K. Dialynas‡1,2, Dimitra Makatsori‡1, Niki Kourmouli§, Panayiotis A. Theodoropoulos§, Kevin McLean¶,Stefan Terjung�, Prim B. Singh**3, and Spyros D. Georgatos‡4

From the ‡Stem Cell and Chromatin Group, Laboratory of Biology, The University of Ioannina, School of Medicine and IoanninaBiomedical Research Institute/Foundation for Research and Technology, 45 110 Ioannina, Greece, the §Department of BasicSciences, University of Crete, School of Medicine, 95 110 Heraklion, Crete, Greece, the ¶Functional Genomics Unit, MoredunResearch Institute, EH26 0PZ Penicuik, Midlothian, Scotland, United Kingdom, the �Advanced Light Microscopy Facility (ALMF),European Molecular Biology Laboratory, 69117 Heidelberg, Germany, and the **Nuclear Reprogramming Laboratory, Departmentof Gene Expression and Development, Roslin Institute, EH25 9PS Edinburgh, Scotland, United Kingdom

We have examined HP1�-chromatin interactions in differentmolecular contexts in vitro and in vivo. Employing purified compo-nents we show that HP1� exhibits selective, stoichiometric, andsalt-resistant binding to recombinant histone H3, associating pri-marily with the helical “histone fold” domain. Furthermore, using“bulk” nucleosomes released byMNase digestion, S-phase extracts,and fragments of peripheral heterochromatin, we demonstrate thatHP1� associatesmore tightlywithdestabilizedordisruptednucleo-somes (H3/H4 subcomplexes) than with intact particles. Westernblotting and mass spectrometry data indicate that HP1�-selectedH3/H4particles and subparticles possess a complex pattern of post-translational modifications but are not particularly enriched inme3K9-H3. Consistent with these results, mapping of HP1� andme3K9-H3 sites in vivo reveals overlapping, yet spatially distinctpatterns,while transient transfection assayswith synchronized cellsshow that stable incorporation of HP1�-gfp into heterochromatinrequires passage through the S-phase. The data amassed challengethe dogma that me3K9H3 is necessary and sufficient for HP1 bind-ing and unveil a new mode of HP1-chromatin interactions.

Histone modifications are thought to provide specific readouts thatare selectively utilized in DNA transactions or chromatin state transi-tions (1). Given the multiplicity of modification sites and the diversechemistries of post-translational modifications, the combinatorial rep-ertoire of this putative “histone code” might have enormous dimen-sions; for instance, methylation of the five lysine residues that are

located at the amino-terminal tail of histone H3 could yield alone over15 � 103 distinct patterns, while “saturation marking” of all lysines,arginines, serines, and threonines that are found in the same regionwould result in�256� 106 combinations. Clearly then, even if 1% of thepredicted patterns were materialized in vivo, this voluminous “instruc-tion manual” could not be functionally interpreted without the aid ofspecific de-coding factors. Consistent with this idea, recent studies haveidentified a set of chromatin-associated proteins that bind specificallymodified histones and could, at least in theory, fulfil such a de-codingrole. As it turns out, these “effector” molecules are often components oflarge enzymatic assemblies and possess specialized modules known asbromo-, tudor-, or chromodomains (2).A classic example of a chromodomain-containing protein is HP1, a

conserved constituent of eukaryotic cells, which, in metazoans, com-prises three distinct variants: �, �, and � (3). HP1� and HP1� are local-ized in compact heterochromatic regions, whileHP1� ismore abundantin euchromatic territories (reviewed in Refs. 4 and 5). All HP1 variantshave the same molecular architecture: they contain an amino-terminalchromodomain (CD),5 an intervening region (“hinge”) and a carboxyl-terminal chromoshadow domain (CSD). The CD is thought to beresponsible for chromatin association, whereas the CSD represents amultipurpose binding platform for nuclear chaperones, remodeling fac-tors, and histone-modifying enzymes (4, 5). Interaction sites are alsoaccommodated in the hinge region, which, in addition, contains a func-tional nuclear localization signal (6).Although the picture is rapidly changing (e.g. Ref. 7), it is still widely

believed that HP1 targets transcriptionally inactive, constitutive hetero-chromatin by binding specifically to histone H3 that has been post-translationally modified (trimethylated) at lysine 9 (me3K9-H3). Thistenet is based on several observations: first, HP1�/HP1� largely co-localize with me3K9-H3 (8, 9) and seem to accumulate in the neighbor-hood of silenced genes (7); in addition, these two variants are completelyde-localized when the enzymes responsible for heterochromatin-spe-cific Lys9 trimethylation (SUV39H1/H2) are genetically ablated (8, 9);finally, me3K9-H3 peptides and native H3 have been reported to bindHP1 in vitro, whereas unmodified peptides and recombinant H3 do not(10–13).

* This work was supported in part by PENED and EPAN grants from the General Secre-tariat of Research and Technology and by a Pythagoras-EPEAK grant from the Ministryof Education (Greece) (to S. D. G.), a core grant from the Biotechnology and BiologicalScience Research Council (United Kingdom) (to P. B. S.), and by a Royal Society (UnitedKingdom) exchange award (to P. B. S. and S. D. G.). The costs of publication of thisarticle were defrayed in part by the payment of page charges. This article must there-fore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734solely to indicate this fact.

This paper is dedicated to Giorgos Mathianakis.� This article was selected as a Paper of the Week.□S The on-line version of this article (available at http://www.jbc.org) contains supple-

mental data and Figs. S1–S3.1 Both authors contributed equally to this work.2 Supported in part by an Ioannina Biomedical Research Institute/Foundation for

Research and Technology-internal fellowship.3 Present address: Division of Tumor Biology, Dept. of Immunology and Cell Biology,

Forschungszentrum Borstel, D-23845 Borstel, Germany.4 To whom correspondence should be addressed: Laboratory of Biology, University of

Ioannina, School of Medicine, Dorouti, 45 110 Ioannina, Greece. Tel.: 302651097565;Fax: 302651097863; E-mail: sgeorgat@cc.uoi.gr.

5 The abbreviations used are: CD, chromodomain; CSD, carboxyl-terminal chro-moshadow domain; FRAP, fluorescence recovery after photobleaching; NG, Nuclearghost; NGE, NG extract; DTT, dithiothreitol; PMSF, phenylmethylsulfonyl fluoride;BrdUrd, bromodeoxyuridine; GST, glutathione S-transferase; MALDI-TOF, matrix-as-sisted laser desorption ionization time-of-flight; LBR, lamin B receptor; D, diffuse; SP,speckled; me1, methylated; me3, trimethylated.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 281, NO. 20, pp. 14350 –14360, May 19, 2006© 2006 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

14350 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

Despite the elegance and simplicity of this model, binary interactionsbetween HP1 and me3K9-H3 do not seem to account for the wholespectrum of HP1-heterochromatin associations that are observed invivo (14). A survey of different organisms, cell types, and chromosomalpreparations shows that the HP1 and me3K9-H3 patterns are notexactly coincident (15, 16), whilemapping of HP1 andme3K9-H3 targetloci inDrosophila reveals the existence of distinct binding sites (17–19).In line with these observations, HP1 seems to dissociate from hetero-chromatin when cellular components unrelated to Suvar3,9 (e.g., theorigin recognition complex protein ORC2 or RNA) are removed (20–24). Last, but not least, in vitro binding of HP1 to non-methylated his-toneH3 and nakedDNAhas been recently claimed (25–27), contradict-ing some of the previous observations.Prompted by current controversies, we decided to study HP1-chro-

matin interactions at different levels of complexity. Purified histoneswere employed to examine the binding properties ofHP1 independentlyof chromatin state and presence of auxiliary factors. Alternatively,“bulk” chromatin, released from nuclei byMNase digestion, was used toinvestigate the association of HP1 with native particles of different sizeand epigenetic status. Finally, chromatin fragments obtained from frac-tionated “nuclear ghost” and S-phase extracts were exploited to assessHP1 interactions with intact core particles and assembly/disassemblyintermediates.Aiming at specific, stoichiometric interactions, we adopted a “�g-

scale” binding assay and analyzed HP1-associated histones by SDS-PAGE-Coomassie Blue staining and mass spectrometry, avoiding pur-posely the use of radiolabeled tracers (in vitro transcribed/translatedmaterial) and resorting to Western blotting only when we wanted to“diagnose” epigeneticmodifications. These in vitro studies were done inparallel with in vivo experiments, employing transient transfectionassays, double-immunolabeling in combination with confocal micros-copy, and fluorescence recovery after photobleaching (FRAP). The datareveal a new mechanism of HP1 binding to histone H3 and stronglysuggest that stable incorporation of this protein in chromatin territoriesoccurs after passage through the S-phase.

MATERIALS AND METHODS

Cells and Antibodies—Turkey erythrocytes were obtained fromwhole blood. HeLa, MCF-7, NRK, and MDCK2 cells were culturedaccording to standard procedures. Anti-acetylated Lys14-H3 and anti-SNF2 polyclonal antibodies were purchased from Upstate Biotechnol-ogy, Lake Placid, NY. The characterization of anti-me3K9-H3 and anti-me3K20-H4 antibodies has been described (15, 28). CREST humanautoimmune serumwas kindly provided byW. C. Earnshaw (Universityof Edinburgh, Edinburgh, UK).

Cell Cycle Arrest—Early S-phase cells were obtained by treatmentwith 5mM hydroxyurea for 24 h. Alternatively, we used 3 �g/ml aphidi-colin for 24 h, 0.5 mM mimosine for 24 h, or a double thymidine block(19 h in 2 mM thymidine, 9-h release, 16 h in 2 mM thymidine).

Cell Fractionation and Preparation of Nuclear Extracts—Nuclearghosts (NGs) were prepared from turkey erythrocyte nuclei, as specified(29). Briefly, after three rounds of DNase I digestion (80�g/ml in 10mM

NaPO4, 2 mM MgCl2, 10% sucrose, 1 mM DTT, 1 mM PMSF/proteaseinhibitors, 15 min, at room temperature, with 200 �g/ml RNase Aincluded in the last digestion step), the resulting NGs were washed inbuffer I (150 mM NaCl, 20 mM Tris-HCl pH 7.5, 250 mM sucrose, 2 mM

MgCl2, 0.1 mM EGTA, 1 mM DTT, 1 mM PMSF), sonicated to inducevesiculation, and stored at �80 °C. Extracts were prepared by thoroughresuspension of sonicated NGs in 300–600mMNaCl, 20mMTris-HCl,pH 7.5, 250mM sucrose, 2 mMMgCl2, 0.1 mM EGTA, 1mMDTT, 1mM

PMSF, protease inhibitors. After ultracentrifugation (200–350,000 � g,30 min, 4 °C), the soluble extracts were collected and either used imme-diately or further fractionated in 10–30% sucrose density gradients(100,000� g, 4 °C, 18 h). For preparation of S-phase extracts, HeLa cellswere synchronized using double thymidine block, collected bytrypsinization, washed three times with isotonic buffer (150 mM NaCl,20 mM Hepes, pH 7.4, 2 mM MgCl2, 1 mM PMSF), and resuspended inbufferA (150mMNaCl, 20mMHepes, pH7.4, 2mMMgCl2, 1mMCaCl2,0.25% Triton X-100, 250 mM sucrose, 1 mMDTT, 1 mM PMSF, 2 �g/mlprotease inhibitors). Cells were treated withMNase (5 units/106 cells or1,500 units/ml of pellet) for 10min at 37 °C, incubated on ice for 10minafter addition of 2 mM CaCl2, and centrifuged at 13,000 � g, for 10 minat 4 °C. The supernatant (F1) was collected, and the salt concentrationwas adjusted to 300 or 600 mM. The pellet was resuspended in ice-cold2 mM EDTA, pH 7.5, and centrifuged as above. The supernatant wascollected and the salt adjusted to 300 or 600 mM (F2). The pellet wasfurther extracted with 300 mM salt buffer (300 mM NaCl, 20 mM Tris-HCl, pH 7.5, 250 mM sucrose, 2 mMMgCl2, 0.1 mM EGTA, 1 mM DTT,1 mM PMSF, protease inhibitors) and 1% Triton X-100, centrifuged at100,000 � g, for 30 min at 4 °C, and the supernatant (F3) collected.Equivalent amounts of fractions F1, F2, and F3 were used for pulldownassays.

Microscopy—For light microscopy, samples were fixed with 1–4%formaldehyde in phosphate-buffered saline, permeabilized with 0.2%Triton X-100, and blocked with 0.5% fish skin gelatin. DNA staining(propidium iodide) and probing with the relevant primary and second-ary antibodies was performed according to Maison et al. (30). BrdUrdstaining was done as specified by manufacturer (Roche DiagnosticsGmbH, Penzberg, Germany).

Pulldown Assays—GST fusion proteins (HP1�, HP1�, and HP1� orGST alone; �15–30 �g) were incubated first for 30 min at room tem-perature with 30 �l of glutathione-agarose beads, “blocked” in 1% fishgelatin in assay buffer. After washing three times with the samemedium, the beads were combined with nuclear extracts and furtherincubated for 1 h at room temperature. The beads were washed sixtimes with buffer, before eluting the bound proteins with hot SDS-sample buffer.

Other Methods—MALDI-TOFmass spectrometry was performed atthe Functional Genomics Unit of Moredun Research Institute, Edin-burgh, UK. Protein bands were digested with R-specific protease. �M(difference betweenmeasured and calculatedmasses) was at the level of1/10,000. Peak assignment was done either manually or using AppliedBiosystems programs. Trypsin digestion experiments were performedusing 100 �l of trypsin cross-linked to agarose beads (19.5 units/ml)incubated with the corresponding protein (recombinant H3) for 0, 2,and 10 min at 4 °C. The reaction was terminated by addition of trypsinand protease inhibitors. The digests were ultracentrifuged at 200–350,000 � g (30 min, 4 °C) and then used for pulldown experiments.SDS-PAGE, Western blotting, column chromatography, and microin-jection were practiced according to established procedures. Morpho-metric analysis and assessment of co-localization of different markerswere performed using stacks of confocal images digitally processed andanalyzed by Image J.

RESULTS

HP1 Binding to Purified Histones and Histone Peptides—To examinethe binding preference of HP1, we first assayed total histones extractedfrom avian erythrocyte nuclei by 0.1 M H2SO4. As shown in Fig. 1A,HP1�-GST precipitated almost exclusively histoneH3. Consistent withthis, HP1� bound stoichiometrically to recombinant histone H3 but

Associations of HP1�

MAY 19, 2006 • VOLUME 281 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14351

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

failed to associate with recombinant histone H4 (Fig. 1A, rec H4 and recH3). Binding to recombinant H3 was tight and easily detectable even at0.75 M salt (Fig. 1B). Since no significant differences were seen whenHP1� or histones were treated with DNase I/RNase A or when stoichi-ometric amounts of plasmid DNA were included in the assay (Fig. 1C),nonspecific interactions mediated by nucleic acid contaminants couldbe safely ruled out.As shown in Fig. 1D, “tail” peptides that represented either the non-

modified or the me3K9-modified NH2-terminal region of H3 exhibitedmarginal binding to HP1� at 0.3 M salt and did not compete with thefull-length protein, even at a 200 molar excess. However, when H3 was

digested with trypsin, a proteolytic product with an apparent Mr of12,000 was detected in the HP1� precipitate (Fig. 1E, CP). On the basisof previously published studies (31–33) and results shown in Fig. 1F, thisfragment can be assigned to the stretch extending between amino acids40–129, i.e. the so-called “histone fold” domain. Thus, at least under invitro conditions, HP1� has the capacity to bind histone H3 independ-ently of post-translational modifications, associating primarily with thecentral, helical part of the molecule.

HP1 Interactions with Fragments of Native Chromatin—To studyHP1 interactions with bulk chromatin, we used as a substrate materialreleased from G0 nuclei after MNase digestion. As shown in Fig. 2A,when the whole digest was used as input, HP1� precipitated stoichio-metric amounts of histone proteins. However, when chromatin wasfractionated on sucrose density gradients (Fig. 2,B andC) and individualfractions assayed, we noticed that HP1� bound almost exclusively tooligonucleosomal arrays (nu�) and did not interact significantly withmononucleosomes (nu1) (Fig. 2D). The higher propensity of HP1� tobind oligonucleosomes was not due to “enrichment” in me3K9 histoneH3, as documented in theWestern blot shown in Fig. 2E. Furthermore,no differences were seenwhen the experiments were done at 0.3 and 0.6M salt (data not shown), a point that will be taken up next.

Continuing these studies, we then examined S-phase chromatin,which is expected to contain a variety of nucleosome assembly/disas-sembly intermediates (34, 35). Cultured cells were arrested at G1/S witha double thymidine block and then released for 7 h to allow progressionto mid-S (Fig. 3A, inset). A “soluble” chromatin fraction (F1) was col-lected after lysis with Triton X-100 and “nicking” with MNase (mildnuclease digestion was deemed necessary to release detached particlestrapped in tangles of non-replicating chromatin). Larger fragments ofeuchromatin (F2) and heterochromatin (F3) were subsequentlyextracted from theTriton-insoluble residue by serial washingwith 2mM

EDTA and 0.3 M NaCl (see diagram in Fig. 3A).

FIGURE 1. HP1 binding to isolated histones and histone peptides. A, material precip-itated by HP1� (lanes 1), or GST (lanes 2), upon co-incubation with a mixture of erythro-cyte histones, recombinant H3 (rec H3), or recombinant H4 (rec H4). (Inp) is 50% of theinput. B, recombinant H3 precipitated by HP1� (lanes 1), or GST (lanes 2), at 0.3, 0.6, and0.75 M salt. Inp is 50% of the input. C, histone H3 precipitated by HP1� (odd-numberedlanes) or GST (even-numbered lanes) after treatment with DNase I/RNase A and in thepresence of plasmid DNA, as indicated. Both recombinant (lanes 1– 4 and 9 and 10) andnative (lanes 5– 8 and 11 and 12) histones were used. D, binding of recombinant H3 toHP1� in the presence of a 200-fold excess of the unmodified tail peptide (lane 3), theme3K9-modified peptide (lane 4), or buffer (lane 5). Lanes 1 and 2 show electrophoreticprofiles of the unmodified and the modified H3 peptide, respectively. HP1�, recombi-nant H3, and co-precipitating peptides (N-tail) are indicated. E, digestion of recombinantH3 with trypsin beads for 0 min (lane 1), 2 min (lane 2), and 10 min (lane 3). Lanes 4 – 6,material precipitated from each of these digests by HP1�. A peptide corresponding tothe “fold” domain of histone H3 (CP) and the intact protein are indicated. Lane on the leftcorresponds to molecular mass markers (in kDa). F, sequences of the CP peptide, asdetected by mass spectrometry. Stretches that are expected to yield distinct peaks (seealso Fig. 6) are in bold, whereas products that were either too small or too large foranalysis by MALDI-TOF are in regular letters. Detected peptides are underlined. The pro-tein has been cleaved with R-specific protease and the data analyzed statistically byMascot Search.

FIGURE 2. HP1 binding to mononucleosomes and oligonucleosomes. A, HP1 bindingto bulk chromatin, released from nuclei after MNase digestion. Lane 1, material precipi-tated by HP1� at 0.3 M salt; lane 2, corresponding GST control. Inp indicates a sample ofthe input material (20%). B, fractionation of the MNase digest by centrifugation in a10 –30% sucrose gradient. Fractions corresponding to mononucleosomes and oligonu-cleosomes are designated (nu1) and (nu�), respectively. C, DNA profile of fractions (nu1)and (nu�). Sizes (in bp) are indicated on the right. D, pulldown assay with selected gradi-ent fractions. In each case Inp indicates the input material (20%) and Bound the HP1�precipitate. E, Western blot with anti-me3K9 antibodies corresponding to the samplesshown in D.

Associations of HP1�

14352 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

When the F1 fraction was used as input and the assay performed at0.3 M salt, HP1� precipitated apparently intact histone octamers (Fig.3B, F1, lane 1). However, when the same experiment was done at 0.6 M

salt, the HP1� precipitate contained predominantly H3/H4 (Fig. 3B, F1,lane 2), suggesting that Triton-soluble particles are labile and probablyrepresent non yet stabilized (nascent) or de-stabilized nucleosomes (forrelevant data, see Ref. 36). Supporting this interpretation, no salt-de-pendent effects were observed with fractions F2 and F3 (Fig. 3B, F2 andF3, compare lanes 1 and 2), which are known to contain stable, deter-gent-insoluble chromatin pieces (i.e. polynucleosomes).

HP1 Interactions with Intact Particles and Subparticles Derived fromHeterochromatin—In combination, the observations presented in Figs.2 and 3 indicated that HP1� does not bind to intact mononucleosomesreleased from G0 chromatin but does bind oligonucleosomes and “salt-labile”, detergent-soluble fragments extracted from S-phase chromatin.Since these preparations represented bulk chromatin, not particularlyenriched in any single histone modification, we wondered what wouldhappen if the input contained native particles that are specifically mod-ified in Lys9-H3. To answer this question, we used fractions derivedfrom peripheral heterochromatin. In a previous study we have shownthat mononucleosomes highly enriched in me3K9 and depleted ofme3K4 can be obtained in high yield by 0.3 MNaCl extraction of nuclearghosts, i.e. G0 nuclei that have been extensively digested with DNase Iand sonicated, to release pieces of heterochromatin firmly attaching to

the nuclear envelope (for procedural details and mass spectrometrydata, see Ref. 29). When such extracts (DNase-NGEs) were co-incu-batedwithHP1� under different ionic conditions, we observed an inter-esting and highly reproducible pattern. At 0.3 M salt, HP1� precipitatedthe four core histones, en blocwith fragments of DNA containing 100–250 bp (Fig. 4, A and B, lanes 1). However, upon closer inspection, onecould easily see that the ratio of the core histones in the HP1� precipi-tate was slightly “skewed” in favor of H3/H4. This “preference” becamepronounced when the assay was executed in 0.6 M salt, at which point

FIGURE 3. HP1 binding to fractions of S-phase chromatin. A, diagram showing thescheme used to fractionate S-phase chromatin. The inset shows HeLa cells stained withBrdUrd (BrdU) and propidium iodide (PI) before synchronization (NT), after thymidineblock (THY), and after release for 7 h in normal medium (R). B, binding of HP1� to nuc-leosomal particles extracted from S-phase cells. Lanes 1 and 2, material associated withHP1� at 0.3 and 0.6 M salt, respectively; lanes 3, GST controls (at 0.3 M salt). Inp indicates20% of the material added in each assay. F1, F2, and F3 correspond to chromatin-solubi-lized sequentially by Triton X-100, EDTA and 0.3 M salt.

FIGURE 4. HP1 binding to disrupted nucleosomes. A, HP1 binding to fragments ofperipheral heterochromatin extracted from nuclease-digested nuclei (nuclear ghosts).Lanes 1 and 1� and 3 and 3�, material precipitated by HP1� at 0.3 and 0.6 M NaCl, respec-tively; lanes 2 and 2� and 4 and 4�, corresponding GST controls. Inp indicates a sample(20%) of the input extract. Group DNase-NGE corresponds to extracts of DNase I-digestednuclei, whereas MNase-NGE to extracts of MNase-digested nuclei (for more details see“Results.”) B, DNA co-precipitating with HP1� and histones in the experiment presentedabove. Lanes are exactly as specified for A. Markers (in bp) are shown on the left.C, Western blotting using anti-me3K9 antibodies. Lanes are exactly as indicated for A.D, fractionation of the nuclear ghost extract (DNase-NGE) in a 10 –30% sucrose gradient.Fractions used in subsequent experiments are indicated by arrows. E and F, DNA profileand Western blot of the fractions indicated in D. Markers (in bp) are shown on the right.G, pulldown assay with the fractions shown above. Odd-numbered lanes correspond tosamples assayed at 0.3 M, while even-numbered lanes correspond to samples assayed at0.6 M salt. H, schematic describing in hypothetical terms the particle types present innuclear ghost extracts. Nucleosomal DNA is represented by broken or wavy lines; H2A-H2B and H3-H4 subcomplexes are shown as semicircles. For further details see “Results.”

Associations of HP1�

MAY 19, 2006 • VOLUME 281 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14353

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

HP1� brought down only H3/H4 subparticles that were essentiallyDNA-free (Fig. 4, A and B, lanes 3).The characteristic pattern was not observed when the input material

contained larger particles. Peripheral heterochromatin extracts con-taining oligonucleosomes rather than mononucleosomes could be eas-ily prepared by digesting the nuclei with MNase instead of DNase I,taking advantage of the fact that the former is much less aggressive anuclease than the latter (37, 38). When these extracts (MNase-NGEs)were tested at different salt conditions, HP1� precipitated equimolaramounts of the four core histones and DNA fragments consisting of150-mer even at 0.6 M salt. This experiment proves that histone octam-ers do not disassemble by a mere salt “jump” from 0.3 M to 0.6 M NaCl,unless nucleosomal DNA has been extensively cut (for further detailssee below). This is not inconsistent with the current literature: thebreakpoint at which H2A/H2B separate from H3/H4 and DNAunwraps from the particle lies somewhere between 0.5 and 1 M NaCl(11, 39–42); however, the exact “transition zone” seems to depend crit-ically on Mg2� and nucleosome concentration (11, 39–42).Since histoneH3 precipitated fromDNase-NGEs at 0.3 and 0.6 M salt

did not differ with regards to Lys9 trimethylation (Fig. 4C), the datapresented in Fig. 4A could be interpreted to mean that under stringentconditions HP1� exhibits preferential binding to H3/H4 subparticles.

That such disrupted nucleosomes pre-existed in the input materialand did not originate from salt-induced disassembly could be confirmedby separating DNase-NGEs on sucrose density gradients at 0.3 M NaCl(Fig. 4, D–F) and assaying different fractions at both 0.3 and 0.6 M salt.As shown in Fig. 4G (lanes 3 and 4), when low density material (e.g. fr7)was used as input, HP1� precipitated exclusively H3/H4 at both saltconcentrations. However, when material collected from the middle/high density zone (e.g. fr12-fr19) was tested in the sameway, the bindingpattern was clearly salt-dependent and resembled that observed withthe whole, non-fractionated DNase-NGEs (compare Fig. 4G, lanes 5–8,with Fig. 3A, lanes 1–4). Evidently, HP1� did not bind linker histones(Fig. 4G, lanes 1 and 2) that were abundantly present in the top fractionsof the gradient (Fig. 4D).

The interpretationwe offer to explain these results is presented in thecartoon of Fig. 4H. We argue that the low density fractions of the gra-dient shown in Fig. 4D contain predominantly disrupted particles (nuD).These probably originate fromnucleosomeswhoseDNAhas been com-pletely digested during chromatin preparation (removal of DNA isknown to weaken H2A/H2B-H3/H4 interactions and favor particle dis-sociation; reviewed in Ref. 43). We further suggest that the middle/highdensity fractions are enriched in intact mononucleosomes (nu1) andparticles that lack various parts of nucleosomal DNA and are partially“open” (nuO). Unlike nu1, nuO can bind HP1� at permissive conditions,similarly to nuD and non-stabilized S-phase intermediates. “Skewing” infavor of H3/H4 is observed at 0.3 M salt (Fig. 4, A and G), becauseoctameric units (nuO), as well as half-particles consisting of H3/H4, areco-precipitated with HP1�. This becomes more pronounced when thesalt is increased, as “unstable” nuO split and are converted to nuD. Sincenu1 contain intact nucleosomal DNA and are not salt-labile, only twoparticle species remain at 0.6 M NaCl: these that bind HP1� (nuD) andthose that do not (nu1).To directly examine whether disrupted particles were more apt to

associate with HP1� than intact mononucleosomes, we took advantageof the fact that LBR, a chromatin-binding protein structurally unrelatedto HP1, has a predilection for intact nucleosomes (Ref. 31; see Fig. S1 insupplemental data). Knowing that, we assessed binding using as input apreparation that contained both fully and partially assembled nucleo-somes (44). Results presented in Fig. 5A make it clear that under the

same (permissive) conditions, and at equivalent protein concentrationLBR “collects” octameric particles, while HP1� selects predominantlyH3/H4 complexes.To further confirm these results, we performed pulldown assays

under limiting conditions. As shown in Fig. 5B, H3/H4 complexes werepreferentially precipitated at 0.3 M salt when the HP1� concentration inthe reactionmixturewas gradually decreased, suggesting selective bind-ing to disrupted particles (nuD). This could also be shown by successivepulldown assays at 0.6 M salt. In that case, HP1� was incubated with afixed amount of DNase-NGEs, as described in the legend to Fig. 4A.After collecting the first precipitate, a new aliquot of the protein wasadded to the supernatant (non-bound fraction) and the procedurerepeated nine more times. As illustrated in Fig. 5C, the extract wasgradually depleted of H3/H4 subparticles, while a considerable amountof core histones, presumably representing intact mononucleosomes(nu1), were left behind (compare lanes Inp and NB).

Finally, to eliminate factors that are peculiar to each chromatin iso-lation-fractionation technique we assessed side by side the bindingproperties of H3/H4 subparticles (e.g. nuD, isolated as in Fig. 4D) withthat of oligonucleosomes (e.g. nu�, isolated as in Fig. 2B). Both speciesassociated nearly stoichiometrically with HP1� (Fig. 5D).

Modification Patterns of HP1-associated Histone H3—To determinethemodification “signatures” of HP1-associated histoneH3, we isolatedH3/H4 subparticles that bind to HP1� at 0.6 M salt (see Fig. 4A) andanalyzed the samples by mass spectrometry. We also examined histoneH3 precipitated in the context of octameric particles (i.e. at 0.3M salt). Inthe latter case, all three HP1 variants (�,�, �) were used in the pulldownexperiments, to identify potential differences in binding preference.Consistentwith the biochemical data described above (Figs. 1–4), the

modification patterns were identical in all histoneH3 samples analyzed.As could be seen in Fig. 6A, methylation of K4 and double acetylation atLys18/Lys23 were not detected. However, affinity-selected H3 exhibitedother interesting features, e.g. trimethylation in Lys79 and (mono)acety-lation in Lys18/Lys23, which are, supposedly, “euchromatic” modifica-tions. Among the species identified were also trimethylated Lys27-H3and methylated Lys36-H3. Of these modifications, trimethylation of

FIGURE 5. Selective binding of HP1 to H3/H4 sub-particles. A, particles “selected” byLBR (lane 1), HP1� (lane 2), and GST (lane 3) from a mixture of fully and partially assemblednucleosomes at 0.3 M salt. Inp represents 20% of the input. B, pulldown assays withvarying amounts of HP1� (decreasing from lane 1 to lane 4) at 0.3 M salt. Inp represents20% of the input extract, which contains nu1, nuO, and nuD. C, histones precipitated byrecombinant HP1� in 10 consecutive pull downs at 0.6 M NaCl (lanes 1–10). Samples ofthe nuclear ghost extract (Inp; 20%) and the non-bound material (NB; 20% of total) areincluded. D, histones co-sedimenting with HP1� when the input (Inp, 20%) containeddisrupted particles (nuD, lane 1), or oligonucleosomes (nu�, lane 2).

Associations of HP1�

14354 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

Lys27 has been associated with Polycomb-mediated gene silencing,while methylation of Lys36 is thought to provide a “transcription ON”signal (for a review, see Ref. 2).The stretch of amino acids 9–17, which contains Lys9 and Lys14, was

rather heavily modified. Since the mass difference corresponding to Lystrimethylation is very close to that of Lys acetylation, a peak at 971.5 Dacould be assigned either to a (dimethylated � acetylated) or to a (di-methylated � trimethylated) peptide. Likewise, another peak at 985.5Da could be attributed either to a twice trimethylated, or to a (trimethyl-ated� acetylated) peptide (Fig. 6B). Since no histonemethyltransferasespecific for Lys14 has been described so far, while both me3K9-H3 andacetylated Lys14-H3 could be identified in HP1-precipitated H3 byWestern blotting (Fig. 6C), we favor the latter interpretation. Be that asit may, the modification patterns detected were not diagnostic and didnot conform to the formula “me3K9/non-acetylated K14/me1K27” thatis thought to be necessary and sufficient for HP1 binding (11, 39–42).

Incorporation of HP1 into Chromatin under in Vivo Conditions—Ourbiochemical studies suggested strongly thatHP1 proteins associate withdifferent chromatin substrates in a manner that depends crucially onphysical state. Since chromatin is a dynamic structure and undergoesmultiple transitions during the cell cycle (43), we predicted that our invitro observations likely had an in vivo correlate. To explore this idea, we

employed a transient transfection system. Two different human celllines (HeLa, MCF-7) and several methods of DNA uptake (microinjec-tion, calcium phosphate treatment, or electroporation) were used inthese experiments. Cells injected or porated with HP1�-gfp plasmidswere examined at early (2–6 h) and late (12–24 h) time points, whilecalcium phosphate-transfected cells were visualized at 12, 24, and 48 h.Irrespective ofmethod, two distinct phenotypeswere always detected

among transfected cells: phenotype “SP” (after “speckled”) included cellsin which HP1�-gfp had accumulated at perinucleolar and peripheralheterochromatin; phenotype “D” (after “diffuse”) represented anothersubpopulation inwhich the gfp-tagged protein did not follow the knownstereotype and was diffusely distributed throughout the nucleoplasm,occasionally forming small granules (Fig. 7A). Interestingly, the relativeproportions of SP and D cells changed with time in a rather orderlyfashion (Fig. 7B): initially, accumulation of HP1�-gfp in heterochro-matic areas was observed only in 30% of the cells; however, this propor-tion increased steadily, reaching a value of �80% after 24 h. Similarresults were obtained with non-human cells (e.g. canine MDCK2 cells,and rat NRK; data not shown).The different patterns observed could be attributed to a variety of

factors, including over-expression and, as a result, aberrant localization.For this reason, we first examinedwhether the distribution of HP1�-gfpcorrelated with fluorescence intensity. When the nuclei of numeroustransfected cells were photographed and categorized, we could notdetect an obvious connection between the overall fluorescence intensityand the specific HP1�-gfp pattern (Fig. 7C). However, it was noticeablethat, on the average, SP cells had slightly larger nuclei than D cells (Fig.7D). Taking this into account, we performed a morphometric study,recording precisely the ratio of fluorescence intensity/nuclear surfacearea (fi/a) in 80 equatorial sections taken at the confocal microscope 6 hpost-transfection. The results confirmed that overexpression does notaccount for the phenotypes observed: on the average, SP cells had only amarginal edge over D cells (differences in fi/a ratio not exceeding 10%),while individual D and SP figures often exhibited exactly the same flu-orescence intensity.We could also confirm in a number of ways that assembly of HP1�-

gfp in this transient transfection system was not “ectopic.” First, HP1�-gfp and endogenous HP1� co-localized to more than 90% (Fig. 8A).Second, as expected from previous studies, HP1�-gfp foci fell wellwithin heterochromatic regions (as defined by accumulation ofme3K9-H3 and me3K20-H4) and outside euchromatic territories (asdefined by SNF2 distribution) (Fig. 8, B and C). However, thoroughinspection of transfected cells with confocal microscopy did reveal sub-tle features of the HP1 assemblies that were not immediately obviousduring the initial screening of the specimens. For instance, whereasHP1�-gfp and endogenousHP1�were both present in heterochromaticfoci, one could distinguish discrete subdomains, in which the relativeproportion of the two proteins varied significantly (note “variegated”foci in the gallery of Fig. 8A). Moreover, it was clear in numerous (xy)and (xz) sections that HP1�-gfp and me3K9-H3 do not exactly colocal-ize (Fig. 8C, arrows). Since HP1�-gfp was capable of targeting hetero-chromatic sites containing endogenous HP1� and exhibited normaldynamics (see below), steric effects due to the gfpmoiety could be safelyruled out. Therefore, the spatially distinct patterns of HP1�-gfp andme3K9-H3 fluorescence could mean that HP1� follows differentassembly pathways, one of which is largely independent of Lys9

trimethylation.This idea was supported by other experiments. For instance, when

human cells were transfected with two HP1� mutants (one possessingonly the CD and the other containing the hinge plus the CSD), or a

FIGURE 6. Modification patterns of HP1-associated histone H3. A, a list of the peaksdetected by mass spectrometry in samples of histone H3 precipitated by HP1�, HP1�,and HP1�. nm, non-modified residues; me, methylated residues; ac, acetylated residues.B, a segment of the mass spectrum depicting the most interesting peaks (those corre-sponding to 971.5 and 985.5 Da are indicated by arrows). C, Western blot of the HP1�-bound histones after probing with anti-me3K9 and anti-acetyl Lys14-H3 antibodies.Lanes correspond to the bound (B) and the non-bound (NB) fraction (100% of the mate-rial has been analyzed). The assays were performed at 0.6 M salt.

Associations of HP1�

MAY 19, 2006 • VOLUME 281 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14355

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

mixture thereof, we obtained exclusively the D-phenotype, irrespectiveof plasmid uptake technique and time course (Fig. 8D). This was rathertelling, because according to current studies (10) the CD-containinggfp-protein should have been able to target me3K9-enriched sites.

Cell Cycle-dependent Assembly and Dynamics—As mentionedabove, the nuclei of D cells were smaller than those of SP cells, suggest-ing cell cycle differences (nuclear size increases visibly after theG1-phase). To explore this further, we co-injected the HP1�-gfp plas-mid and the base analog Cy3-dUTP and examined the cells 6 h later. Asshown in Fig. 9A, cells exhibiting a D pattern were in most of the casesCy3-dUTP-negative and, therefore, non-S. When diffuse fluorescenceco-existed with a Cy3-dUTP-positive staining, the distribution of thebase analog was typical of an early-S cell (euchromatic replication pat-tern). However, in all cases where large HP1�-gfp blocks and peripheralfluorescence were detected (SP-phenotype), the cells were either Cy3-dUTP-positive (late-S, “heterochromatic” replication pattern) or earlypost-mitotic (i.e. paired, at telophase or early G1). Cy3-dUTP andHP1�-gfp spots in late-S cells were not always coincident (Fig. 9A,mid-dle panels), because the base analog could be apparently incorporatedinto replicating DNA immediately after injection, whereas the HP1�-gfp protein was synthesized with a lag.We also monitored transfected cells that had incorporated BrdUrd

(see Fig. S2 in supplemental data). In this case, the samples were exam-ined 18 h after transfection and 30 min after incubation with this baseanalog. Although a smaller number of figures could be scored using thismethod, the results were essentially the same as with the Cy3-dUTPexperiments presented above. The combined observations strongly sug-gested that stable incorporation of HP1� in heterochromatic domainsoccurs in late S-phase and at the end of mitosis. Consistent with thelatter point, staining of transfected cells with anti-centromere (CREST)antibodies revealed reversible dissociation of HP1�-gfp from pericen-tromeric heterochromatin during cell division (Fig. 9B), a propertyshared with endogenous HP1 (45, 46).

To investigate the same problem from a different angle, we trans-fected cells that had been cultured either in normal medium or in thepresence of cell cycle-arresting agents. As illustrated in Fig. 9C (upperpanels), HP1�-gfp assembled normally in the control cells, following theusual 24-h schedule. However, when the cells were blocked at theS-phase (by either hydroxyurea, thymidine, or aphidicolin), or in G1 (bymimosine), HP1�-gfp assembly at heterochromatic sites was inhibitedand 90% of the cells exhibited a D pattern. This dramatic effect was nota consequence of “global” unbalance, because transfection of unsyn-chronised and cell cycle-arrested cells by control plasmids (Rab7-gfpand CD39 mutant-gfp) yielded the expected pattern in both cases (Fig.9C,middle and lower panels). In line with these observations, when thecells were released from hydroxyurea block, the proportion of figuresthat exhibited the typical, SP pattern became increasingly higher and 8 hafter entering the S-phase themajority had recruited HP1� into hetero-chromatin (Fig. 9,D and E). These results lead us to conclude that stableincorporation of HP1� at heterochromatic sites requires passagethrough the S-phase.Recently published FRAP data on HP1-gfp transfected cells have

shown that all forms ofHP1-gfp exhibit rapid dynamics, with the bulk ofthe fluorescence returning to the bleached area within seconds (36, 45,47, 48). To find out whether this holds in our experimental system, weperformed FRAP and inverse FRAP studies on transiently transfectedHeLa cells. The results (presented in Fig. S3 of supplemental data) weresimilar to those reported previously in refs (36, 45, 47, 48), showing rapidrecovery of euchromatin-bound HP1�-gfp and slightly slower, but stillvery fast, kinetics of the heterochromatin-bound tracer. Consistentwiththe observations presented in Fig. 9, the mobility of HP1�-gfp inhudroxyurea-arrested cells was comparable with that of diffuse HP1�-gfp in non-arrested cells butmuch lower than the reported values of freegfp (36, 45, 47, 48), which would correspond to freely diffusible, non-boundmaterial. From these observationswe conclude that the transient

FIGURE 7. De novo assembly of HP1 in tran-siently transfected human cells. A, representa-tive phenotypes after transfection of HeLa cellswith HP1�-gfp and counter-staining with pro-pidium iodide (PI). The columns correspond tosequential confocal sections. B, morphometricdata from different transfection experiments,using HeLa and MCF-7 cells. The proportion of fig-ures exhibiting diffuse or focal patterns aredepicted in the histogram. n is the number of cellsscored. C, relative HP1�-gfp expression levels incells exhibiting either diffuse or focal nuclear fluo-rescence 6 –12 h after transfection. For furtherexplanations see “Results.” D, relative nuclear sizein transfected cells exhibiting a diffuse or a speck-led phenotype. n is the number of cells scored. Forfurther explanations see “Results.”

Associations of HP1�

14356 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

system employed in this study does not differ significantly from thepreviously used models.As puzzling as it might seem, the fact that HP1�-gfp exhibits rapid

exchange, while its accumulation and stable incorporation in hetero-chromatic territories proceeds in a cell cycle-dependent fashion, can berationally explained. It is possible that a subpopulation of HP1 mole-cules, which escapes detection by FRAP, is recruited to chromatin onlywhen specific “windows of opportunity” arise. This small subpopulationmay serve as a seed for furtherHP1 assembly. A speculativemodel basedon this hypothesis and accommodating all of our in vivo and in vitroobservations is presented in Fig. 10.

DISCUSSION

me3Lys9 as an Exclusive HP1 Binding Motif—While mis-localizationof HP1 in Suvar3,9 null cells provides compelling evidence thatme3K9-H3 is critical for HP1 targeting to heterochromatin, other inter-pretations are still open for discussion. Knock-out of the two Suvar3,9genes, apart from damping Lys9 trimethylation, also causes a variety ofdownstream effects, such as abolishment of Lys20 trimethylation in his-tone H4, mis-targeting of the DNA methylase Dnmt3b and increase ofLys27 trimethylation in histone H3 (28, 49, 50). Thus, it is hard to dis-tinguish whether HP1 is mis-targeted because heterochromatin archi-

tecture is globally altered or because there is a lack of appropriate bind-ing sites.In vitro studies bearing on the question of whether or not me3K9

constitutes the sole binding site for HP1 sometimes yield surprisingresults; for instance, while H3 peptides containing me3K9 bind HP1with �M affinity (10, 51), oligonucleosomal arrays reconstituted fromrecombinant (i.e. unmodified) histones seem to bind with nM affinity(25). Low affinity binding of the H3 tail is consistent with the fact thatsynthetic peptides representing amino acids 1–15 bind weakly to HP1but do not compete effectively with the intact protein, even at a 200-foldmolar excess. This could only occur if HP1 had a higher affinity for thehistone fold region of H3, as previously suggested by Nielsen and others(26, 50, 52). In line with this interpretation, we have found that HP1binds to recombinant (i.e. non-modified) and trypsinized (i.e. tailless)H3, while HP1-selected particles of native heterochromatin contain acomplex pattern of modifications that do not conform to the formula“me3K9/non-acetylated K14/me1K27” (50, 52). This is not the first timethat the currently accepted histone code “rules” are violated: modifica-tion of Lys9 (trimethylation), Ser10 (phosphorylation), and Lys14 (acety-lation) have been detected recently on the same H3 molecule underin vivo conditions (53).

Multiple Pathways of HP1 Assembly—Previous work with stablytransfected cells has established that at steady-state HP1�/HP1�-gfplocalize in large heterochromatic blocks around nucleoli and in densechromatin masses located at the nuclear periphery (Ref. 45 and refer-ences therein). However, when de novo assembly of HP1 proteins isinterrogated using a transient transfection system (Ref. 36, 45, 47, 48and this study), or microinjection of purified HP1 proteins (46), morethan one phenotype is usually observed. In the latter case, accumulationof exogenous proteins in nucleoplasmic foci occurs with a considerablelag and is not completed until 24 h post-injection (i.e. approximately theduration of the complete cell cycle).It is conceivable that aberrantHP1 assemblymight occur as a result of

physical trauma (e.g. injection), chemical stress (e.g. calcium phosphatetreatment), or metabolic unbalance (e.g. overloading the cells with “for-eign” proteins). Aware of these limitations, in this study we have exam-ined cells that express different levels of wild type or mutant HP1 pro-teins, experimented with alternative transfection protocols, and testeddifferent cellular models. Irrespective of method, we always observedtwo distinct HP1�-gfp patterns in human cells: in one group of cells thefluorescent protein was scattered throughout the nucleoplasm in theform of “granules,” or tiny foci, whereas in another subpopulation it wasfully incorporated into large blocks of heterochromatin. This pheno-typic variation was not due to nuclear anomalies that can be detected byDAPI/PI and anti-histone, anti-HP1, anti-SNF-2, or anti-nuclear enve-lope antibody staining (Fig. 8).6 Moreover, the relative proportions ofthe two phenotypes evolved in an orderly fashion upon progression ofthe cell cycle. These observations and the data amassed from in vitrostudies led us to conclude that stable incorporation of HP1 proteins intoperipheral and perinucleolar heterochromatin takes place only whenand where nucleosome structure is altered.The in vivo circumstances under which this might occur are well

defined: partial or complete dissociation of the core particle could occurduring transcription-associated remodeling, recombination, histonevariant exchange, or even steady-state chromatin dynamics (54–56).Nevertheless, by and large, nucleosome disassembly and reassembly isknown to occur during DNA replication. Supporting a replication-de-

6 G. K. Dialynas, D. Makatsori, N. Kourmouli, P. A. Theodoropoulos, K. McLean, S. Terjung,P. B. Singh, and S. D. Georgatos, unpublished observations.

FIGURE 8. Targeting of HP1-gfp to heterochromatic foci. A, counter-staining of HP1�-gfp-transfected cells by anti-HP1� antibodies. The gallery on the right shows (at highercontrast and zoom) the spectrum of color patterns detected in different heterochro-matic foci. For further comments see text. B and C, relative distribution of HP1�-gfp andSNF-2 (euchromatic marker) or me3K9-H3 and me3K20-H4 (heterochromatic markers). xyand xz sections are shown. Arrows indicate foci that are singly decorated (either red orgreen). nu denotes the area of the nucleolus. Other designations are as in previous fig-ures. D, the same experiment as in A, using two HP1� deletion mutants: 1–70 (CD-gfp)and 69 –185 (�CD-gfp).

Associations of HP1�

MAY 19, 2006 • VOLUME 281 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14357

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

pendent pathway, labeling of transfected cells with Cy3-dUTP orBrdUrd revealed that de novo assembly of HP1� requires passagethrough the S-phase. Early after transfection, the majority (over 70%) ofthe cells that can recruit HP1�-gfp into peripheral and perinucleolarheterochromatin are in the late S-phase and exhibit a typical “hetero-chromatic/late replication” pattern. Conversely, virtually all of the cellsthat display diffuse HP1�-gfp fluorescence are either non-S or at theearly S-phase. A diffuse pattern has never been detected in late S-phasecells, while, upon G1 or S-phase arrest, deposition of HP1� to hetero-chromatic blocks is inhibited.

State and Microscopic Species of HP1—The observations reportedhere may appear surprising in view of FRAP studies showing rapidexchange of heterochromatin-bound HP1-gfp in mammalian and yeastcells (Refs. 36, 45, and 47 and this report). However, this discrepancy canbe explained if we take into account that often the recovery of HP1-gfpis not complete, with fluorescence reaching only 70–80% of the pre-

bleaching value. It is possible that heterochromatin-associated HP1comprises several kinetically (and perhaps chemically) distinct species,one of which is highly immobile and serves a “structural” role. Oppositeto loosely associatedmaterial held in place byweak interactions (36, 57),such integrated HP1 molecules would not dissociate from heterochro-matin, unless the octamer structure is disrupted. Support for this idea isprovided by a recent study (48) where several kinetically distinct formsof HP1 were identified. In this report, the relative amount of immobilemolecules correlated with the chromatin condensation state, mountingtomore than 44% in the condensed chromatin of transcriptionally silentcells.Another interpretation, which unifies data obtained in several differ-

ent laboratories, could be that steady-state dynamics and stable incor-poration of HP1 proteins in human cells represent mechanistically dis-tinct processes. Apart from nucleosome “opening”, incorporation ofnewly synthesized HP1 into chromatin may in fact require siRNA, cell

FIGURE 9. Cell cycle dependence of HP1 assem-bly. A, staining patterns of transfected cells 6 hafter co-injection of a HP1�-gfp plasmid and Cy3-dUTP into the cytoplasm. The proportions of eachphenotype are given on the left. n is the number ofcells scored. B, relative distribution of HP1�-gfpand centromeric markers (CREST antigens) inmitotic cells. The right-hand panel in the blowup isrotated in comparison with the original image C.Distribution of HP1�-gfp, Rab7-gfp, and a CD39mutant-gfp 24 h post-transfection in cells culturedin normal media (NT) or media that containedaphidicolin (AP), hydroxyurea (HU), thymidine(THY), or mimosine (MIM). D, resumption of HP1�-gfp assembly after release from hydroxyureablock. Representative phenotypes 8 h after trans-fer to normal medium are shown. The blowupshows parts of the cells that are numbered athigher contrast and zoom. E, morphometric datafrom the experiment shown in C. n is the numberof cells scored. Light gray columns, D cells; darkgray columns, SP cells.

Associations of HP1�

14358 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

cycle-specific modification/“licensing”, or chaperoning by a specificS-phase factor.In Fig. 10 we propose that HP1 binds initially to a variety of dispersed

loci (high affinity/low capacity binding, no exchange), accessing thehistone fold region of H3 that becomes exposed in regions of activetranscription, histone variant exchange, or replication. Once an initial“nucleus” is assembled, HP1may spread to non-replicating heterochro-matin by binding to adjacent, clustered me3K9 sites (low affinity/highcapacity binding, rapid exchange). Obviously, if such determinants donot exist in the immediate neighborhood (e.g. euchromatin), only a traceamount of HP1 would remain associated with the initial sites, yieldingan indistinct signal in indirect immunofluorescence assays and creatingthe impression of an exclusively heterochromatic localization. Ourfuture studies are directed toward rigorously testing this workinghypothesis.

Acknowledgments—Morphological work was performed at the Confocal andVideo Microscopy Unit of the University of Ioannina and at Advanced LightMicroscopy Facility (European Molecular Biology Laboratory). We are grate-ful to Anastasia Politou, Panos Kouklis, Hara Polioudaki, and Carol Murphyfor intellectual contributions and materials.

REFERENCES1. Jenuwein, T., and Allis, C. D. (2001) Science 293, 1074–10802. Daniel, J. A., Pray-Grant, M. G., and Grant, P. A. (2005) Cell Cycle 4, 919–9263. Singh, P. B.,Miller, J. R., Pearce, J., Kothary, R., Burton, R.D., Paro, R., James, T. C., and

Gaunt, S. J. (1991) Nucleic Acids Res. 19, 789–7944. Brasher, S. V., Smith, B. O., Fogh, R. H., Nietlispach, D., Thiru, A., Nielsen, P. R.,

Broadhurst, R. W., Ball, L. J., Murzina, N. V., and Laue, E. D. (2000) EMBO J. 19,

1587–15975. Cowieson,N. P., Partridge, J. F., Allshire, R. C., andMcLaughlin, P. J. (2000)Curr. Biol.

10, 517–5256. Wang, G., Ma, A., Chow, C. M., Horsley, D., Brown, N. R., Cowell, I. G., and Singh,

P. B. (2000)Mol. Cell. Biol. 20, 6970–69837. Litt, M. D., Simpson, M., Gaszner, M., Allis, C. D., and Felsenfeld, G. (2001) Science

293, 2453–24558. Peters, A. H., O’Carroll, D., Scherthan, H., Mechtler, K., Sauer, S., Schofer, C., Wei-

poltshammer, K., Pagani, M., Lachner, M., Kohlmaier, A., Opravil, S., Doyle, M.,Sibilia, M., and Jenuwein, T. (2001) Cell 107, 323–337

9. Schotta, G., Ebert, A., Krauss, V., Fischer, A., Hoffmann, J., Rea, S., Jenuwein, T., Dorn,R., and Reuter, G. (2002) EMBO J. 21, 1121–1131

10. Jacobs, S. A., and Khorasanizadeh, S. (2002) Science 295, 2080–208311. Bannister, A. J., Zegerman, P., Partridge, J. F., Miska, E. A., Thomas, J. O., Allshire,

R. C., and Kouzarides, T. (2001) Nature 410, 120–12412. Lachner,M., O’Carroll, D., Rea, S., Mechtler, K., and Jenuwein, T. (2001)Nature 410,

116–12013. Nakayama, J., Rice, J. C., Strahl, B. D., Allis, C. D., andGrewal, S. I. (2001) Science 292,

110–11314. Ahmad, K. (2004)Mol. Cell 15, 494–49515. Cowell, I. G., Aucott, R., Mahadevaiah, S. K., Burgoyne, P. S., Huskisson, N., Bon-

giorni, S., Prantera,G., Fanti, L., Pimpinelli, S.,Wu, R., Gilbert, D.M., Shi,W., Fundele,R., Morrison, H., Jeppesen, P., and Singh, P. B. (2002) Chromosoma 111, 22–36

16. Ringrose, L., Ehret, H., and Paro, R. (2004)Mol. Cell 16, 641–65317. Greil, F., van der Kraan, I., Delrow, J., Smothers, J. F., deWit, E., Bussemaker, H. J., van

Driel, R., Henikoff, S., and van Steensel, B. (2003) Genes Dev. 17, 2825–283818. Danzer, J. R., and Wallrath, L. L. (2004) Development (Camb.) 131, 3571–358019. Vakoc, C. R., Mandat, S. A., Olenchock, B. A., and Blobel, G. A. (2005)Mol. Cell 19,

381–39120. Huang, D. W., Fanti, L., Pak, D. T., Botchan, M. R., Pimpinelli, S., and Kellum, R.

(1998) J. Cell Biol. 142, 307–31821. Prasanth, S. G., Prasanth, K. V., Siddiqui, K., Spector, D. L., and Stillman, B. (2004)

EMBO J. 23, 2651–266322. Maison, C., Bailly, D., Peters, A. H., Quivy, J. P., Roche, D., Taddei, A., Lachner, M.,

Jenuwein, T., and Almouzni, G. (2002) Nat. Genet. 30, 329–33423. Muchardt, C., Guilleme,M., Seeler, J. S., Trouche,D., Dejean,A., andYaniv,M. (2002)

EMBO Rep. 3, 975–98124. Quivy, J. P., Roche, D., Kirschner, D., Tagami, H., Nakatani, Y., and Almouzni, G.

(2004) EMBO J. 23, 3516–352625. Fan, J. Y., Rangasamy, D., Luger, K., and Tremethick, D. J. (2004) Mol. Cell 16,

655–66126. Nielsen, A. L., Oulad-Abdelghani, M., Ortiz, J. A., Remboutsika, E., Chambon, P., and

Losson, R. (2001)Mol. Cell 7, 729–73927. Perrini, B., Piacentini, L., Fanti, L., Altieri, F., Chichiarelli, S., Berloco, M., Turano, C.,

Ferraro, A., and Pimpinelli, S. (2004)Mol. Cell 15, 467–47628. Kourmouli, N., Jeppesen, P., Mahadevhaiah, S., Burgoyne, P., Wu, R., Gilbert, D. M.,

Bongiorni, S., Prantera, G., Fanti, L., Pimpinelli, S., Shi, W., Fundele, R., and Singh,P. B. (2004) J. Cell Sci. 117, 2491–2501

29. Makatsori, D., Kourmouli, N., Polioudaki, H., Shultz, L. D., McLean, K., Theodoro-poulos, P. A., Singh, P. B., andGeorgatos, S. D. (2004) J. Biol. Chem. 279, 25567–25573

30. Maison, C., Horstmann, H., and Georgatos, S. D. (1993) J. Cell Biol. 123, 1491–150531. Bohm, L., and Crane-Robinson, C. (1984) Biosci. Rep. 4, 365–38632. Ausio, J., Dong, F., and van Holde, K. E. (1989) J. Mol. Biol. 206, 451–46333. Fletcher, T. M., and Hansen, J. C. (1995) J. Biol. Chem. 270, 25359–2536234. Chang, L., Loranger, S. S., Mizzen, C., Ernst, S. G., Allis, C. D., and Annunziato, A. T.

(1997) Biochemistry 36, 469–48035. Gruss, C., Wu, J., Koller, T., and Sogo, J. M. (1993) EMBO J. 12, 4533–454536. Cheutin, T., Gorski, S. A., May, K. M., Singh, P. B., and Misteli, T. (2004) Mol. Cell.

Biol. 24, 3157–316737. Lee, M. S., and Garrard, W. T. (1991) EMBO J. 10, 607–61538. Cartwright, I. L., and Elgin, S. C. (1986)Mol. Cell. Biol. 6, 779–79139. Yager, T. D., and van Holde, K. E. (1984) J. Biol. Chem. 259, 4212–422240. Oohara, I., and Wada, A. (1987) J. Mol. Biol. 196, 389–39741. Oohara, I., and Wada, A. (1987) J. Mol. Biol. 196, 399–41142. Park, Y. J., Dyer, P. N., Tremethick, D. J., and Luger, K. (2004) J. Biol. Chem. 279,

24274–2428243. Luger, K., and Hansen, J. C. (2005) Curr. Opin. Struct. Biol. 15, 188–19644. Oudet, P., Germond, J. E., Sures,M., Gallwitz, D., Bellard,M., andChambon, P. (1978)

Cold Spring Harb. Symp. Quant. Biol. 42, 287–30045. Cheutin, T., McNairn, A. J., Jenuwein, T., Gilbert, D. M., Singh, P. B., and Misteli, T.

(2003) Science 299, 721–72546. Kourmouli, N., Theodoropoulos, P. A., Dialynas, G., Bakou, A., Politou, A. S., Cowell,

I. G., Singh, P. B., and Georgatos, S. D. (2000) EMBO J. 19, 6558–656847. Festenstein, R., Pagakis, S. N., Hiragami, K., Lyon, D., Verreault, A., Sekkali, B., and

FIGURE 10. Pathways of HP1 incorporation into chromatin. This highly speculativemodel claims that assembly of HP1 proteins can proceed in two different ways. A, stablebinding to dispersed chromatin sites (via the histone fold) that become available underspecific circumstances and phases of the cell cycle. Nucleosomal DNA is represented bywavy lines; H2A-H2B and H3-H4 subcomplexes are shown as shaded or open semicircles,respectively. B, loose binding to clustered low affinity sites (through me3K9). For moredetails see “Discussion.”

Associations of HP1�

MAY 19, 2006 • VOLUME 281 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14359

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

Kioussis, D. (2003) Science 299, 719–72148. Schmiedeberg, L., Weisshart, K., Diekmann, S., Meyer Zu Hoerste, G., and Hemmer-

ich, P. (2004)Mol. Biol. Cell 15, 2819–283349. Lehnertz, B., Ueda, Y., Derijck, A. A., Braunschweig, U., Perez-Burgos, L., Kubicek, S.,

Chen, T., Li, E., Jenuwein, T., and Peters, A. H. (2003) Curr. Biol. 13, 1192–120050. Schotta, G., Lachner, M., Sarma, K., Ebert, A., Sengupta, R., Reuter, G., Reinberg, D.,

and Jenuwein, T. (2004) Genes Dev. 18, 1251–126251. Jacobs, S. A., Taverna, S. D., Zhang, Y., Briggs, S. D., Li, J., Eissenberg, J. C., Allis, C. D.,

and Khorasanizadeh, S. (2001) EMBO J. 20, 5232–524152. Fischle, W., Wang, Y., and Allis, C. D. (2003) Nature 425, 475–47953. Mateescu, B., England, P., Halgand, F., Yaniv, M., and Muchardt, C. (2004) EMBO

Rep. 5, 490–49654. Flaus, A., and Owen-Hughes, T. (2004) Curr. Opin. Genet. Dev. 14, 165–17355. Korber, P., and Horz, W. (2004) J. Biol. Chem. 279, 35113–3512056. Kimura, H., and Cook, P. R. (2001) J. Cell Biol. 153, 1341–135357. Singh, P. B., and Georgatos, S. D. (2002) J. Struct. Biol. 140, 10–16

Associations of HP1�

14360 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 20 • MAY 19, 2006

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from

GeorgatosTheodoropoulos, Kevin McLean, Stefan Terjung, Prim B. Singh and Spyros D.

George K. Dialynas, Dimitra Makatsori, Niki Kourmouli, Panayiotis A. into HeterochromatinβIncorporation of HP1

Methylation-independent Binding to Histone H3 and Cell Cycle-dependent

doi: 10.1074/jbc.M600558200 originally published online March 18, 20062006, 281:14350-14360.J. Biol. Chem. 

  10.1074/jbc.M600558200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2006/03/21/M600558200.DC1

  http://www.jbc.org/content/281/20/14350.full.html#ref-list-1

This article cites 57 references, 30 of which can be accessed free at

by guest on January 9, 2021http://w

ww

.jbc.org/D

ownloaded from